The human retinoblastoma gene (RB-1) is a well-established tumor suppressor. Somatic RB-1 inactivation is observed in approximately one third of all human tumors. Individuals carrying germline RB-1 mutations develop retinoblastoma with near complete penetrance and they are also highly predisposed to develop osteosarcomas. The retinoblastoma protein (pRB) inhibits cellular proliferation by binding to the E2F transcription factors and thereby blocks the activation of genes encoding essential components of the cell division machinery. pRB has also been shown to bind transcription factors that are master regulators of differentiation. Together, these observations suggest that pRB acts to coordinate cell cycle arrest and terminal differentiation. The analysis of mutant mouse models and cell lines has been used to gain insight into the roles of pRB and E2F4 in vivo. First, these studies show that E2F4-loss suppresses the development of pRB-deficient tumors. Molecular studies suggest a hypothesis for the underlying basis of E2F4's oncogenic activity. Experiments in aim 1 will directly test this hypothesis. Second, these studies show that pRB and E2F4 are required for the appropriate development of numerous tissues. In particular they reveal important roles for both pRB and E2F4 in the development of bone arising via either intramembranous or endochondral ossification. Experiments in aim 2 will continue to investigate the mechanisms by which pRB and E2F4 contribute to differentiation with particular emphasis on bone development. In Aim 3, we will use osteoblast-specific Rb mouse models to further investigate pRB's role in osteogenesis and osteosarcoma formation. [unreadable] [unreadable] [unreadable]